[2605.07662] Direction-Preserving Number Representations

-->

Computer Science > Machine Learning

arXiv:2605.07662 (cs)

[Submitted on 8 May 2026]

Title:Direction-Preserving Number Representations

Authors:Bardia Zadeh, George A. Constantinides<br>View a PDF of the paper titled Direction-Preserving Number Representations, by Bardia Zadeh and George A. Constantinides

View PDF<br>HTML (experimental)

Abstract:Low-precision number formats are widely used in modern machine learning systems due to their efficiency. Accurate direction representation is key to the accuracy of vector operations. This work precisely explores the extent to which the direction of a vector can be represented by selecting its scalar elements from a common finite alphabet of a given size. This is standard practice in machine learning, where low-precision significands may be narrow-width floating-point or integer values. A geometric framework is introduced for analyzing the directional coverage of such product-structured codes.

This work analytically quantifies the suboptimality gap between such product-structured codes and spherical codes for the vector as a whole, in both low and asymptotically high dimensions. Furthermore, within the product code class, it is proven that the standard formats of two's complement, fixed-point, and floating-point are suboptimal, again with quantified gap, pointing to the potential to develop new scalar number formats.

Such scalar alphabets are numerically optimized across multiple block dimensions for directional coverage, including the dimension used in NVIDIA's NVFP4 format. Experimental results are presented comparing the performance of standard formats and the optimized alphabet. We find that for four bits, NVIDIA's choice of E2M1 closely approximates the optimized alphabet, providing a geometric explanation for its strong performance in low-precision machine learning workloads and an analytical understanding of the link between that superiority and block size.

We provide open-source formal proofs in Lean for the theorems in this work, along with the experimental code and the optimized alphabets obtained.

Comments:<br>9 pages excluding appendices and references, 18 in total. 5 figures

Subjects:

Machine Learning (cs.LG); Numerical Analysis (math.NA)

ACM classes:<br>G.1.0

Cite as:<br>arXiv:2605.07662 [cs.LG]

(or<br>arXiv:2605.07662v1 [cs.LG] for this version)

https://doi.org/10.48550/arXiv.2605.07662

Focus to learn more

arXiv-issued DOI via DataCite (pending registration)

Submission history<br>From: Bardia Zadeh [view email]<br>[v1]<br>Fri, 8 May 2026 12:32:33 UTC (799 KB)

Full-text links:<br>Access Paper:

View a PDF of the paper titled Direction-Preserving Number Representations, by Bardia Zadeh and George A. Constantinides<br>View PDF<br>HTML (experimental)<br>TeX Source

view license

Current browse context:

cs.LG

next >

new<br>recent<br>| 2026-05

Change to browse by:

cs<br>cs.NA<br>math<br>math.NA

References & Citations

NASA ADS<br>Google Scholar

Semantic Scholar

export BibTeX citation<br>Loading...

BibTeX formatted citation

&times;

loading...

Data provided by:

Bookmark

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)

Related Papers

Recommenders and Search Tools

Link to Influence Flower

Influence Flower (What are Influence Flowers?)

Core recommender toggle

CORE Recommender (What is CORE?)

IArxiv recommender toggle

IArxiv Recommender<br>(What is IArxiv?)

Author

Venue

Institution

Topic

About arXivLabs

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs .

Which authors of this paper are endorsers? |<br>Disable MathJax (What is MathJax?)